Member Junction
    Preparing search index...

    Interface DuplicateReasoningCandidateVerdict

    The reasoner's verdict for ONE candidate, judged independently against the source. A matched set can mix true duplicates and false positives (vector search returns the top-K neighbors), so each candidate gets its own recommendation/confidence/reasoning — the detector stamps the per-candidate verdict onto that candidate's match row, never a blanket set-level verdict.

    interface DuplicateReasoningCandidateVerdict {
        AIPromptRunID?: string;
        Confidence: number;
        Reasoning: string;
        Recommendation: DuplicateReasoningRecommendation;
        RecordID: string;
    }
    Index

    Properties

    AIPromptRunID? Confidence Reasoning Recommendation RecordID

    Properties

    AIPromptRunID?: string

    The AI Prompt Run that produced THIS verdict, when it is not the set's DuplicateReasoningOutput.AIPromptRunID. DecisionThenPrompt sets it on each candidate its decision dropped, so that candidate's match row points at the decision run rather than at a prompt run that never saw it. Omit it when the set's run produced the verdict.

    Confidence: number

    This candidate's confidence (0–1), or null when the reasoner didn't return one.

    Reasoning: string

    Short rationale specific to THIS candidate.

    Recommendation: DuplicateReasoningRecommendation

    This candidate's verdict vs. the source record.

    RecordID: string

    The candidate record id (must match one of the input candidates' RecordID).

    Settings

    Member Visibility

    On This Page

    Properties